請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/63965
標題: | 應用零力矩點追蹤法於自動導引自行車之平衡控制 Control of Autonomous Bicycle System with ZMP Tracking |
作者: | Mei-Wen Fang 方玫文 |
指導教授: | 周瑞仁(Jui-Jen Chou) |
關鍵字: | 自動導引自行車,平衡控制,零力矩點,追蹤控制, Autonomous Bicycle,Balance Control,Zero Moment Point (ZMP),Tracking Control, |
出版年 : | 2012 |
學位: | 碩士 |
摘要: | 本論文旨在發展自動導引自行車系統,並引入零力矩點(Zero Moment Point, ZMP)觀念,設計出自動導引自行車之平衡控制器。本研究所發展之自動導引自行車系統包含三個主要部分:一般自行車改裝而成的平台、控制器及用來偵測自行車傾斜角與速度的感測器。控制器會先接收感測器的資料,經過處理後轉換成自行車的傾斜角與速度並輸入至控制器中,再經轉向與前進機構模組,控制自行車龍頭轉角來達到平衡。在此系統中,用來控制自行車平衡的變數為龍頭轉角,從零力矩點的路徑可以看出自行車是否平衡,以自行車兩輪接地點的連線作為支撐多邊形,將零力矩點控制在此連線上,便能使自行車達到平衡。本研究設計出一零力矩點追蹤控制器,控制龍頭轉角造成零力矩點的位移,進而使得自行車維達到平衡。根據模擬結果,自行車不管是走直線還是圓形軌跡時,零力矩點的位置皆在0附近振盪,顯示零力矩點追蹤控制能使得自行車達到平衡。在實地測試中,自行車在9秒內可以自我平衡,造成模擬與實測結果不同的原因如下:機械系統無法預期的振動、路面狀況不平均以及模擬時的動力學模型過於簡化等等,因此未來將裝設平衡質量塊於自動導引自行車系統上,用以控制質心位置,增加控制變數,幫助平衡控制。 This study modifies a common bicycle into an autonomous with the proposed zero moment point (ZMP) tracking approach for balancing. The autonomous bicycle system in this study consists of three major parts: a platform modified from a general bicycle, controller, and sensors. The controller will receive the data from sensors and convert them into the lean angle and the forward velocity of the bicycle. Then, the controller will drive the motor on the handlebar in order to keep balance by controlling the steer angle. The control variable used in this study is the steer angle on the handlebar. The ZMP trajectory represents whether the bicycle is balance or not. The objective for balance control is to make the position of the ZMP within the supporting polygon which is the line between the grounding point of the front and the grounding point of the rear wheel. The simulation results show that the position of ZMP always converges to 0 m. It indicates that ZMP tracking control strategy can keep the bicycle balance when the bicycle moves along a straight or circular trajectory. And once disturbance occurs in the system, the lean angle will oscillate. In the field test, the bicycle will be balance in a period of time about 9 s. The reasons why the results between simulation and field test are so different might be the unpredictable vibration from mechanism, the uneven ground and the oversimplified dynamic model. Thus, one control input may not be enough to keep the autonomous bicycle system balance. The balancers could be added for one more control variable in the future work. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/63965 |
全文授權: | 有償授權 |
顯示於系所單位: | 生物機電工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-101-1.pdf 目前未授權公開取用 | 2.49 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。